Up-down, one-zero, forth-back, all-nothing: whatever the way is in which you consider them the goal is the alternation between only two conditions.
By concentrating on the hardware we want to realize it: so where to start from?
High and low signals as referred to a ground one are furnished by every transistor and this means it's our first brick in the wall: let's say a very common one is enough for the purpose (for example the cheap BCxxx family bjts).
What kind of alternation is possible in output?
Not as the value of signals in itself but as the activation and control of the on-off.
We have a single generic bjt and we know that the output will give respectively high and low voltage if the base will have more or less than the diode voltage drop between base and emitter.
By having a npn type and by simply referring the emitter to the ground (and by fixing the collector to positive supply) the base needs more than generally 0.5V to active the transistor: classical example!
While every transistor has a working region where the amplification prevails over the on-off status we indeed need just these last two.
Whatever we want to do with a bjt then an input signal is required and that's what we must think of.
So we have our transistor and a pilot we must individuate.
The question now arises: what kind of drive can it be? Manual or automatic?
Surely automatic: manual electronics sounds quite confusing and stupid but elementary questions can help us to move step by step.
If we need automation again we look for... electronic devices: which could be the proper one?
Yes, boring answer but a new bjt.
Maybe it's possible to put it in quite as the same configuration as the first one with the collector of the second driving the base of first, but this just would add a device more without solving the matter because nothing would change indeed.
So we need a control for the first bjt which can guarantee automation: a new bjt is the quickest and cheapest solution but if it is so surely not in the way just described because we just would move the problem from the first to the second bjt.
Now we have two transistors: what we need is an alternation between their outputs so that while the first is high/low the second is low/high.
Furthermore by looking at the single bjt it has low voltage at the collector if the base has enough "high" voltage on itself.
This is the key!
The collector of every bjt drives the base of the other one.
But even we have found a possible solution to the very first problem, the automation, how can we be sure that it will work?
In other words: most likely two identical bjts are used for a basic experiment like this.
This leads to the issue of symmetry by which we cannot predict:
- if the circuit will work;
- which bjt will saturate first and which will interdict instead.
While the issue number 2 is not a real one the first is indeed. Will it work?
Remember that the map never is the territory: this means that what you see on the paper or on this screen represents the circuit but it's not that.
Nothing can be equal to another: two bjts with the same sign are physically very similar but not perfectly the same. This guarantees the needed asymmetry and unbalance to make the circuit start.
The bjts exchange continuously their outputs in an automatic fashion (until the power supply is present).
That's what we want!
Now that the working principle has been shown as valid another question comes up: what to do with a circuit like this?
Next time we'll see what and how to get it.